Microarchitectural analysis of image quality assessment algorithms

Thien D. Phan, Siddharth K. Shah, Damon M. Chandler, Sohum Sohoni

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Algorithms for image quality assessment (IQA) aim to predict the qualities of images in a manner that agrees with subjective quality ratings. Over the last several decades, the major impetus in IQA research hasfocused on improving predictive performance; very few studies have focused on analyzing and improving the runtime performance of IQA algorithms. This paper is the first to examine IQA algorithms from the perspective of their interaction with the underlying hardware and microarchitectural resources, and to perform a systematic performance analysis using state-of-the-art tools and techniques from other computing disciplines. We implemented four popular full-reference IQA algorithms (most apparent distortion, multiscale structural similarity, visual information fidelity, and visual signal-to-noise ratio) and two no-reference algorithms (blind image integrity notator using DCT statistics and blind/referenceless image spatial quality evaluator) in C++ based on the code provided by their respective authors. We then conducted a hotspot analysis to identify sections of code that were performance bottlenecks and performed microarchitectural analysis to identify the underlying causes for these bottlenecks. Despite the fact that all six algorithms share common algorithmic operations (e.g., filterbanks and statistical computations), our results revealed that different IQA algorithms overwhelm different microarchitectural resources and give rise to different types of bottlenecks. Based on these results, we propose microarchitectural-conscious coding techniques and custom hardware recommendations for performance improvement.

Original languageEnglish (US)
Article number013030
JournalJournal of Electronic Imaging
Volume23
Issue number1
DOIs
StatePublished - Jan 2014

Fingerprint

Image quality
Hardware
resources
hardware
visual signals
discrete cosine transform
ratings
Signal to noise ratio
recommendations
integrity
Statistics
coding
signal to noise ratios
statistics
causes

Keywords

  • image quality assessment
  • microarchitectural analysis
  • performance analysis
  • Vtune

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Science Applications
  • Atomic and Molecular Physics, and Optics

Cite this

Microarchitectural analysis of image quality assessment algorithms. / Phan, Thien D.; Shah, Siddharth K.; Chandler, Damon M.; Sohoni, Sohum.

In: Journal of Electronic Imaging, Vol. 23, No. 1, 013030, 01.2014.

Research output: Contribution to journalArticle

Phan, Thien D. ; Shah, Siddharth K. ; Chandler, Damon M. ; Sohoni, Sohum. / Microarchitectural analysis of image quality assessment algorithms. In: Journal of Electronic Imaging. 2014 ; Vol. 23, No. 1.
@article{8581c3622d9d4c359b3bf0f2b6c058e0,
title = "Microarchitectural analysis of image quality assessment algorithms",
abstract = "Algorithms for image quality assessment (IQA) aim to predict the qualities of images in a manner that agrees with subjective quality ratings. Over the last several decades, the major impetus in IQA research hasfocused on improving predictive performance; very few studies have focused on analyzing and improving the runtime performance of IQA algorithms. This paper is the first to examine IQA algorithms from the perspective of their interaction with the underlying hardware and microarchitectural resources, and to perform a systematic performance analysis using state-of-the-art tools and techniques from other computing disciplines. We implemented four popular full-reference IQA algorithms (most apparent distortion, multiscale structural similarity, visual information fidelity, and visual signal-to-noise ratio) and two no-reference algorithms (blind image integrity notator using DCT statistics and blind/referenceless image spatial quality evaluator) in C++ based on the code provided by their respective authors. We then conducted a hotspot analysis to identify sections of code that were performance bottlenecks and performed microarchitectural analysis to identify the underlying causes for these bottlenecks. Despite the fact that all six algorithms share common algorithmic operations (e.g., filterbanks and statistical computations), our results revealed that different IQA algorithms overwhelm different microarchitectural resources and give rise to different types of bottlenecks. Based on these results, we propose microarchitectural-conscious coding techniques and custom hardware recommendations for performance improvement.",
keywords = "image quality assessment, microarchitectural analysis, performance analysis, Vtune",
author = "Phan, {Thien D.} and Shah, {Siddharth K.} and Chandler, {Damon M.} and Sohum Sohoni",
year = "2014",
month = "1",
doi = "10.1117/1.JEI.23.1.013030",
language = "English (US)",
volume = "23",
journal = "Journal of Electronic Imaging",
issn = "1017-9909",
publisher = "SPIE",
number = "1",

}

TY - JOUR

T1 - Microarchitectural analysis of image quality assessment algorithms

AU - Phan, Thien D.

AU - Shah, Siddharth K.

AU - Chandler, Damon M.

AU - Sohoni, Sohum

PY - 2014/1

Y1 - 2014/1

N2 - Algorithms for image quality assessment (IQA) aim to predict the qualities of images in a manner that agrees with subjective quality ratings. Over the last several decades, the major impetus in IQA research hasfocused on improving predictive performance; very few studies have focused on analyzing and improving the runtime performance of IQA algorithms. This paper is the first to examine IQA algorithms from the perspective of their interaction with the underlying hardware and microarchitectural resources, and to perform a systematic performance analysis using state-of-the-art tools and techniques from other computing disciplines. We implemented four popular full-reference IQA algorithms (most apparent distortion, multiscale structural similarity, visual information fidelity, and visual signal-to-noise ratio) and two no-reference algorithms (blind image integrity notator using DCT statistics and blind/referenceless image spatial quality evaluator) in C++ based on the code provided by their respective authors. We then conducted a hotspot analysis to identify sections of code that were performance bottlenecks and performed microarchitectural analysis to identify the underlying causes for these bottlenecks. Despite the fact that all six algorithms share common algorithmic operations (e.g., filterbanks and statistical computations), our results revealed that different IQA algorithms overwhelm different microarchitectural resources and give rise to different types of bottlenecks. Based on these results, we propose microarchitectural-conscious coding techniques and custom hardware recommendations for performance improvement.

AB - Algorithms for image quality assessment (IQA) aim to predict the qualities of images in a manner that agrees with subjective quality ratings. Over the last several decades, the major impetus in IQA research hasfocused on improving predictive performance; very few studies have focused on analyzing and improving the runtime performance of IQA algorithms. This paper is the first to examine IQA algorithms from the perspective of their interaction with the underlying hardware and microarchitectural resources, and to perform a systematic performance analysis using state-of-the-art tools and techniques from other computing disciplines. We implemented four popular full-reference IQA algorithms (most apparent distortion, multiscale structural similarity, visual information fidelity, and visual signal-to-noise ratio) and two no-reference algorithms (blind image integrity notator using DCT statistics and blind/referenceless image spatial quality evaluator) in C++ based on the code provided by their respective authors. We then conducted a hotspot analysis to identify sections of code that were performance bottlenecks and performed microarchitectural analysis to identify the underlying causes for these bottlenecks. Despite the fact that all six algorithms share common algorithmic operations (e.g., filterbanks and statistical computations), our results revealed that different IQA algorithms overwhelm different microarchitectural resources and give rise to different types of bottlenecks. Based on these results, we propose microarchitectural-conscious coding techniques and custom hardware recommendations for performance improvement.

KW - image quality assessment

KW - microarchitectural analysis

KW - performance analysis

KW - Vtune

UR - http://www.scopus.com/inward/record.url?scp=84897789650&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84897789650&partnerID=8YFLogxK

U2 - 10.1117/1.JEI.23.1.013030

DO - 10.1117/1.JEI.23.1.013030

M3 - Article

AN - SCOPUS:84897789650

VL - 23

JO - Journal of Electronic Imaging

JF - Journal of Electronic Imaging

SN - 1017-9909

IS - 1

M1 - 013030

ER -